The problem with inpatient units…

 

Why are many psychiatric units in the United States such miserable places?  That question came up today on Twitter and there was a consensus by the responders.  It is a chronic question that comes up episodically and there are never any good formulations or solutions. I started working on an inpatient unit in 1988 after three years as the medical director of a community mental health center. At the mental health center, I travelled twice a week to an inpatient unit in a small town where I provided the only psychiatric coverage. Without those visits the inpatient unit would have closed. The new position was at an acute care hospital that accepted all of the emergency psychiatric admissions on the east side of St. Paul, Minnesota.  I was on the unit that accepted the most aggressive patients triaged through the emergency department. Over the next 22 years, a number of factors came into play that made that job impossible to do and resulted in my resignation and moving on to an outpatient job. What follows are my observations about what went wrong.     

1:  Management is strictly on a financial basis with minimal to no psychiatric input and no consideration of quality care.  That means administration typically has no expertise in managing the environmental aspects of care apart from blaming inpatient psychiatrists for any complications that occur. The most glaring deficiency is management of violence and aggression.

When I first started out – there was a psychiatrist who headed the department and set all of the administrative policies. There was a business manager who reported to the head of the department. With the advent of managed care, financial managers replaced psychiatrists as department heads and set administrative policy.  The only variation on that theme is a psychiatrist who carries out administrative decisions from the managed care company administration. The expectation is that the psychiatrists working on inpatient units have minimal to no input on administrative decisions that affect them. There is no discussion of the multiple failed administrative policies from business administrators.

2: Financial management dictates that the admission indication and reason for ongoing care is dangerousness loosely defined as a danger to self or others.  Reviewers aligned with the financial interests of the insurance company make this determination using proprietary guidelines by looking at documentation.  At their discretion they can stop payment for any patient who they determine is not dangerous or suicidal enough to be treated on an inpatient unit. That patient is often immediately discharged.

The clearest sign of failed policy from financial administrators is the current standard for inpatient care. That indication is dangerousness. That means a reviewer can say at any time that a patient will no longer be funded because they are no longer dangerous. This criterion is problematic at many levels. First, it is an inappropriate admission standard that makes it more difficult to assess people in the emergency department. Most people in need of psychiatric admission are in distress but not dangerous. It is not appropriate to turn them away if nothing has been done to alleviate their distress.  Second, dangerousness is stigmatizing and perpetuates the myth that people with psychiatric problems are dangerous. Third, there is no objective way to draw a clear line on a day-to-day basis in order to make a rational discharge decision.

3:  As a direct result of #1; aggressive patients are often triaged to the 5-10% of community hospitals in each state that might be able to contain aggression.

This only applies to states with multiple psychiatric hospitals and in some states that is not true.  Even in states with multiple community hospitals, only a minority of those will have psychiatric units. A select few will admit and treat highly aggressive patients. The reason again is financial. It requires specialized and more intense staffing that costs money.

4:  Length of stay (LOS) is short (3-5 days) to optimize profits.

One of the most perverse incentives are DRG payments. The theory is that the average cost and LOS for a specific diagnosis can be estimated by a group of experts. To financial managers that means, the patient must leave by that duration or less and less is much better. During my tenure in acute care reviewers would call me demanding to know “where is the dangerousness?” that necessitated ongoing inpatient care. Carefully explaining that the patient was not stable enough to function outside of a hospital did not count.  As time went by and managed care companies acquired hospitals this review process was internalized. Inpatient psychiatrists now faced case managers in their team meeting who were basically acting like external reviewers. That impacted not only patient care but the morale and enthusiasm of the inpatient team.

5:  The units are managed to keep all of the beds full irrespective of patient need and there are no private rooms.  This often leads to very incompatible roommate and one of them wanting to leave as a result.  The ability to admit patients is often out of the control of the psychiatric staff and is run by administrators.

Since all inpatient psychiatric beds are rationed in the US and kept at an artificially low aggregate number, these beds are at a premium. In any large hospital the emergency department, the consultation liaison teams, and psychiatric outpatient clinicians are all competing for bed space.  From the minute inpatient psychiatrists arrive in the morning they are pressured to discharge people.  The triage system for admissions is often out of control of the psychiatrists. That results in room mate mismatches and patients not being admitted to their desired specialty units. In both of those situations the inpatient staff and psychiatrists have to address the resulting complaints from patients and families including frequent demands for discharge because of these problems.

6:  Patients are discharged before they are stable to optimize profits.

Severe psychiatric problems rarely respond adequately to treatment in 3-5 days. No medication or psychosocial therapy works that fast. In order to meet the artificial time constraints people are treated aggressively with medications – increasing the risk of side effects.  The ability of the patient to care for themselves in a stable environment is less of a priority.

7:  Many inpatient environments are markedly deficient relative to medical/surgical units (less modern, poor air quality, more crowding, different food service)

This may be changing to some extent with the continued closure of inpatient units. Many of them are dated facilities.  In hospitals where medical surgical patients have private rooms that may not exist on psychiatric units.  In hospitals where there is an ala carte food service for medical surgical patients those choices may not exist on inpatient psychiatric units.  There are many rationalizations for these discrepancies, but when you see the glaring deficiencies in person there is clearly a lack of equal treatment.

In addition to the lack of privacy, practically all acute care units in the US are locked. That certainly reduces the elopement risk and may be necessary from a legal standpoint for involuntary patients, but it is possible to have more liberal policies and allow people off the ward for exercise and passes with their family or friends.  Some research suggests that people may do better on an unlocked unit. The overriding financial oversight comes in to play - with many companies saying that if a person doesn't need to be on a locked ward they don't need to be in a hospital.  Another variation on the dangerousness theme. 

8:  Follow up care is typically lacking in availability and intensity.

For a lot of people, quality inpatient assessment and treatment is their one good shot at stabilization and adequate care. There are many people who have severe mood disorders, bipolar disorder, episodes of psychosis, and postpartum mental illness who have never been stabilized on an outpatient basis. Many have been ill for decades.  Adequate inpatient care can make a significant difference but it will not happen in the span of 3-5 days.  Once adequate care has been established, follow up care is a problem. It is more of a problem if the patient is forced to leave before they are stabilized.

9:  Some units have a disproportionate number of involuntary patients undergoing civil commitment. If committed they may face a very long LOS waiting for transfer to a state hospital in a unit that was not designed for long term care.

The most obvious deficiencies of an inpatient unit come into the light when a patient ends up stranded there for a month or two. They start to experience the cramped quarters and lack of leisure time activity as imprisonment. There has been no work done on how to redesign units for people who have to remain there for extended periods.

10:  Even though substance use disorders are a common comorbidity – they are often seen by the insurance company as a reason for immediate discharge from a psychiatric unit, even when relapse is imminent, it is a life-threatening problem, and no residential beds for the substance use disorder are available.

Insurance company reviewers often insist that patients with severe depression and alcoholism or some other substance abuse problem be discharged the next day. That can even occur if the patient was exhibiting suicidal behavior while intoxicated.  Appropriate detoxification and adequate treatment were not a priority – only the reviewer’s idea that the directly observed suicidal behavior was due to acute intoxication. Most inpatient units do not have immediate access to substance use treatment facilities and it is imperative that these patients are detoxed and stabilized prior to discharge. Business and financial pressure backs up all the way through the psychiatric unit to the emergency department where the message becomes – “people with substance use disorders should not be admitted to psychiatric units.”  This can result in high-risk home detox scenarios and continued relapse with less chance of recovery.  Some counites have "non-medical detox" that patients are transferred to.  They are sent back to the hospital in the event that they have continued significant detox symptoms and may be admitted to a medical service or intensive car unit at that time. 

11:  There is often minimal to no contact with the outpatient staff who were treating the patient prior to admission.

Many outpatient psychiatrists are very cynical about inpatient care. First, they have no control over admissions. They may know inpatient colleagues but realize that it is futile to call them in order to admit one of their patients. They have to tell the patient to go to the emergency department and get assessed for admission. Second assuming that goes well – inpatient staff often do not have the time or energy to consult with outpatient docs about the plan. Finally, they receive many of their patients back who have not improved, are still in crisis, but are now taking higher doses of medication. They typically do not get discharge summaries or other paperwork form the hospital including the discharge medications. 

12:  There is often minimal communication with the family and federal privacy regulations are often given as a reason.

Acute inpatient care is often associated with a family crisis and family members want communication with inpatient staff and the inpatient psychiatrist. Work intensity on the inpatient unit along with staff burnout often results in either a lack of communication or a perceived lack of caring by the family. That can add more conflict to the treatment environment.

13:  The psychiatrists working in these settings have an intense work load and get minimal administrative support. In many cases there is a policing attitude on the part of administrators rather than an affiliative effort.  The psychiatrists are policed on the basis of productivity, LOS, and complications – none of which are under their control.  Staff splitting often occurs because of siloed administration that is commonly used by administration to elicit criticism of specific staff psychiatrists.

Instead of being treated like valuable experts with acknowledged expertise, inpatient psychiatrists are treated like production workers. Administrative staff make decisions that lead to the environment seriously deteriorating and often manage that by becoming more authoritarian and rigid.

14:  Medical coverage is not standardized and emergency department triage is often not enough.

Medical coverage varies greatly depending on the hospital and staff availability. Psychiatrists may not ever touch a patient in some settings or in the case of my inpatient unit – they may be responsible for the complete medical and psychiatric care of the patient.  In some settings there are free standing psychiatric hospitals where ill patients have to be sent by ambulance to an emergency department. In other hospitals there is complete access to all medical and surgical specialties.  In recent years another managed care innovation – the hospitalist has come to inpatient psychiatrist units. That basically means the same psychiatrist works 7 days shift on and 7 days off. Medical coverage is still contingent on local conventions. I have not seen it formally studied, but interviewing Internal Medicine hospitalists left me with the impression that cognitive performance dropped off significantly after 5 days.

Whoever is working the acute care units as a psychiatrist the risk for unrecognized physical illness and destabilized medical problems is always very high. In a chaotic, stressful, unpredictable environment a psychiatrist needs to be at the top of his or her game.

15:  There is intense regulatory interference at all levels.

It is often not obvious that all of the factors I am mentioning here are the direct result of government intervention. The federal government invented the rationed managed care system and early in this century turned the reins over to the insurance industry. It is the single largest conflict of interest interfering with quality care in psychiatry today.  Managed care alone is responsible for many inpatient psychiatric units closing. State sponsored units are rationed on the same principles by human services departments. Both have resulted in a large influx of psychiatric patients into jails where most people do not receive adequate care. Further initiatives like regulating the number of ligature points on an inpatient unit have resulted in further unit closures.

16:  Staff turnover:

It takes a mature and often experienced person to work on an inpatient psychiatry unit – irrespective of their profession. The best inpatient units are held together by a team of psychiatrists, nursing staff, social workers, and occupational therapists. I am convinced that I have worked with some of the best folks from all of those professions. But being the best and being mature enough to be empathic with a unit full of people in extreme distress is not enough. The staff have to be supported and given what they need to be successful. Without that support crises start to happen among the staff. How does that look?  It looks like a social worker who has spent all day on the phone calling 25 nursing homes in order to get a patient placed and being told that they are not doing enough and need to work on placing other patients.  It looks like nursing staff having complex patients taking care of too many patients with high acuity and complicated medical problems with not enough staffing. It looks like nursing assistants being falsely accused of wrongdoing and not being supported.  It looks like various staff members experiencing homicidal threats and nobody knowing what to do about it. Those are just a few examples of what leads to staff turnover.

The staff I worked with knew that we were short of resources. They did everything they could to make the environment more supportive for patients and families. At the Christmas Holiday the occupational therapists would organize a celebration and every patient there got a present and was able to participate. Nursing staff organized a used clothes closet so that patients could be resupplied with clothing if necessary. In some cases we raised cash and transportation on the spot for patients who were leaving abruptly, had no way to get back home, and had no money to buy food.  The inpatient staff is a significant human resource but they can’t compensate for decades of rationing and the irrational polices that play out on their units every day.

17.  Competing forces that increase length of stay that are never addressed by managed care companies:

There are many. The most obvious are probate court polices that affect patients being treated on an involuntary status. Any probate court procedure adds about 2 weeks to the length of stay in the place where I worked.  During that time the patient had no obligation to follow treatment recommendations. That could allow any insurance to refuse payment based on the fact no treatment (apart from containment and psychosocial therapies) was being given.  That creates a number of pressures from administrators and an associated bed shortage. If civil commitment does occur that patient may be waiting for weeks to months for transfer to a state hospital. A more proactive approach in this situation would be to do the hearings on an outpatient basis in the context of community treatment.  I never saw that happen.

Many patients need a therapeutic environment to be discharged to.  They are either homeless or not able to function well enough for independent living. The responsibility of insurance and managed care companies ends at the hospital door. If the inpatient staff cannot find a suitable county or charity funded setting many of these patient are discharged to the street.

Even standard discharge planning to an outpatient clinic can be a problem. Many organizations use a guideline that the patient must be seen in clinic 1-2 weeks post discharge. It is difficult if not impossible to get those appointments even if the inpatient unit and outpatient clinic are in the same organization.  In some cases the appointments are months out with no flexibility in the system to accommodate discharged patients.

All of the factors prolonging inpatient stays by delaying treatment or discharge magnify the pressure on inpatient staff.  Ineffective administrators who cannot negotiate contracts or other arrangements with these outside sources of inpatient utilization transfer that burden directly to the inpatient staff.  The only way to compensate is greater patient turnover and more admissions.  That typically is not possible and the inpatient staff are the obvious scapegoats.

18. Lower reimbursement for equivalent service.

In large metropolitan hospitals psychiatry is an invaluable service in terms of patient flow and discharge planning. Patients with overdoses on medical units and various injuries associated with their psychiatric diagnosis on surgical units – need to be rapidly assessed and transferred or discharged from those primary admitting services.  The emergency department needs to admit psychiatric emergencies to inpatient units. These processes are critical to the function of large hospitals.  Despite that fact, psychiatry is reimbursed at much lower levels for the equivalent amount of care provided by other services. This is an artifact of the long standing carve-out mentality of managed care companies.  In the 1980s they made a decision that psychiatric services were not like the rest of medicine and could be paid for by a separate and lower level of reimbursement. Some of my friends in other specialties, know this and they know that in a hospital setting the high margin services (generally proceduralists) transfer at least part of their profit to cover psychiatric services.  This could all be avoided with equitable reimbursement. Without it funding depends on this transfer of funds and generating as much turnover as possible on the inpatient units.

19:  Psychiatric units in hospitals are the only specialty services that are supposed to be all things to all people.

Most specialists have the luxury of admitting people with a fairly well-defined set of problems. Even if the people are diverse – their problems are not and that specialty service is set up to focus on that set of problems. In the case of inpatient psychiatric units – those rules no longer apply. If the patient has a significant medical or surgical problem and a significant psychiatric problem and the staff psychiatrist has no input into the admission decision – that patient may be admitted to psychiatry. As a result, there are a large group of patients on any unit with significant medical problems that are often acute and need close monitoring. Those problems can interfere with both the patient’s ability to participate in any available programming and also make is difficult to assess any treatment progress focused on their primary psychiatric disorder. The array of these problems can range from acute delirium to a terminal illness requiring intensive nursing care. Since psychiatric units are rarely designed, equipped or staffed to provide this level of care these situations place additional stress on the inpatient environment.  Managed care companies may deny reimbursement for this care on the basis that “the patient should be on a medical unit”.  But of course the medical unit sent the patient in the first place.

20:  Decades of admission avoidance has led to a non-functional admission procedure that is focused on hospital administration needs over outpatient staff and patient needs.

Many outpatient psychiatrists have complained to me over the years that it is impossible to get their patients admitted on a timely basis. On the inpatient side it makes complete sense since the inpatient units are managed to maintain full capacity, there is a chronic bed shortage, and the admissions are not in control of the inpatient psychiatrists. That means the only practical way to get a patient admitted is to send them to the emergency department.  That is true even if the outpatient psychiatrist has consulted with inpatient staff who agree with the admission.  The backlog in the EDs is legendary and there are rules in lace to send the patient to a remote hospital even if that hospital is hundreds of miles away.  There are very few people who want to be voluntarily admitted to a psychiatric unit and even fewer who want to be sent to a remote hospital. 

This conflict plays out in other ways.  In the case of patients with severe depression requiring electroconvulsive therapy (ECT) - they typically cannot be directly admitted and may have to go through the emergency department.  Patients with complicated detoxification related problems - like benzodiazepine detoxification prior to surgery with an associated severe psychiatric problem may not be admitted at all.  There are frequent conflicts about admission and discharge times, because the inpatient staff may end up working long hours (12-13/day) indefinitely due to the timing of the admissions and discharges. In some cases, a hospital may close down their bed capacity and divert all of their admissions to a nearby hospital to avoid this problem.  

21:  Admission Avoidance: This has always been a goal of managed care organizations on both the psychiatric services and medical side of the operation.  There has been a long series of interventions to try to compensate for what amounts to a lack of service and spin it in the most positive light.  About 25 years ago in the New England Journal of Medicine there was an article describing what were essentially crisis units that were supposed to divert potentially short stay psychiatric admissions and house them in a less intensive settings with psychiatric services.  Many counties have this kind of service that is paid for by the county so the cost has been shifted away from managed care companies or federal payers.  I recently attended a conference on a “new” model where a large open hospitable room and psychiatric services are provided. Each patient gets their own lounge chair (the photos I saw showed gerichairs).  There were no beds on the unit. Patients were expected to sleep in those chairs if they had to stay overnight.  Nobody on a 72 hour hold or requiring any significant degree of medical care would be admitted to this unit.  The expectation is that most people would be discharged in about 6-8 hours.  The only real difference from the ED is that patients had more immediate access to psychiatry staff and were not just sitting there waiting to be seen at the next transfer. I suppose some might see this as an innovation. I don’t think you can focus on what is needed on an inpatient unit and what those patients need if you are constantly focused on an artificial admission avoidance concept and putting resources into that.  If anything, it suggests that there are not enough staff and resources on inpatient units.

22.  There is a lack of collaboration with outpatient staff:  Good inpatient care proceeds from the assumption that the main focus of treatment is with the primary psychiatrist or treatment team. For me that attitude goes back to an attending physician I worked with as an intern on an Internal Medicine rotation. He let us know about the term “local MD” and why that was a pejorative. He pointed out that it was arrogance and assumed that the inpatient team who had brief contact with the patient knew more about the care of that person than the outpatient physician.  I did not have enough experience at the time to know one way or the other, but over the years have developed a nuanced view of the problem. But I have no doubt that the inpatient process needs to support outpatient care and that unilateral plans from the inpatient side are by definition suboptimal.

By more nuanced there are a number of reasons for a lack of communication. The only acceptable reasons are that the patient does not have outpatient care, the patient refuses to consent to the communication, or the outpatient physician or their proxy cannot be contacted with a good faith effort. Being on both ends of that call - a good faith effort to me means leaving a cell phone number with the message to “call me at any time.”  I have found that effort is required in an era of overproduction and no set times in the outpatient clinic for necessary phone calls.

In addition to the outpatient psychiatrist, consultants also need to be contacted. I have found that direct communication with the patients cardiologist, endocrinologist, nephrologist, primary care physician, and neurologist is necessary. In fact, there are cases where I do not make any changes to the patient’s medications until I have talked with one of these specialists.

In terms of specific outpatient care, a lot of history needs to be reviewed in the case of complex care.  The outpatient clinic can more efficiently send the records after a brief call. What the outpatient psychiatrist wants to see happen and the endpoint of inpatient care are very important areas that need to be covered. On occasion, the patient expresses dissatisfaction with outpatient care and that conversation can occur in a way that does not split care providers.  For example, one common scenario is the patient with a first manic episode after being treated for years for depression in the outpatient clinic.  A neutral discussion of the difficulty of making a bipolar disorder without a clear manic episode may facilitate transition back to the outpatient psychiatrist.  These problems highlight inpatient psychiatrists needing to maintain a realistic outlook on what has been done and what can possibly be done in the future. 

23:  All of the above factors translate to a chaotic and poorly run inpatient units.  There is no overall clinical guidance because it is typically taken away from psychiatrists and placed with administrators who clearly know nothing about inpatient psychiatry.  

Many inpatient units are nerve wracking places. The first order of business for me after a team meeting was to address as many crises on the unit as possible.  That could include agitated and aggressive patients, patients actively harming themselves, patients refusing medical care for a life-threatening illness, patients refusing surgical care for an obvious problem, and instability due to detoxification from alcohol or benzodiazepines. By addressing these crises, I always hoped to bring a measure of comfort and reassurance to the patient and everyone else who was distressed. I hoped to bring the noise level down. I hoped to have all of the biohazardous material cleaned up.  It is without a doubt a very tough job – made tougher by the fact that you only have the illusion of control. The people really responsible for this bedlam are out of touch. I actually had an administrator tell me to imagine that there was a firewall between me and the administrators who made all of the decisions affecting me, my staff, and the patients. That firewall was there to block my input and the input of my colleagues.

I had planned to do inpatient psychiatry until I retired, but I could not take it anymore. The interpersonal dimension was the most draining. Rather than dwell on that I often think about a deluxe psychiatric hospital that I visited instead. Several years out of residency, I was invited down to this campus by the former chief resident from the program I graduated from. It was a modern campus connected by broad boardwalks running to the compass points. My friend’s office was modern, open and airy. He told me about all of the services and activities available to his patients including excellent cuisine in the cafeteria. At the time the length of stay at his hospital was 2-3 months.  He had no concerns that his patients were unstable at the time of discharge and described none of the stressors that were impacting me on a daily basis. He had set office hours and left at a predictable time every day.  In the subsequent blur of my inpatient tenure, I never found out what happened to this hospital. My suspicion is that managed care eventually shut them down.

I don’t believe for a second that psychiatric inpatient units need to be miserable places that patients and their families want to avoid. I don’t believe for a second that they can’t be therapeutic and stimulating for the dedicated staff that work there.    

But that transformation clearly can’t happen if it is run by business administrators empowered by government edicts.

 

George Dawson, MD, DFAPA

Supplementary 1:

Almost exactly 10 years ago, I had an interview about my thoughts on managed care and psychiatry published in the MetroDocs periodical.  You can read it here but it will probably require adjusting the screen view.

Supplementary 2:

 I have also been interviewed on this theme by Awais Aftab, MD for his series Conversations in Critical Psychiatry.  You can read that interview at the following link.

The Bureaucratic Takeover of American Psychiatry